SPACES aims to present conceptual frameworks
and empirical studies on economic action in
spatial perspective to a wider audience. The
interest is to provide a forum for discussion and
debate in relational economic geography.
Editor Harald Bathelt
Managing Editors Heiner Depner
Katrin Griebel
ISSN 1612-0205 (Printausgabe)
ISSN 1612-8974 (Internetausgabe)
© Copyright Economic Geography, Faculty of Geography,
Philipps-University of Marburg, Germany
Heiner Depner and Harald Bathelt
Cluster Growth and Institutional Barriers: The Development of the Automobile Industry Cluster in Shanghai, P.R. China
SPACES 2003-09
Authors’ Information Heiner Depner, Faculty of Geography, Philipps-University of
Marburg, Deutschhausstraße 10, 35032 Marburg, Germany,
E-Mail: [email protected]
Harald Bathelt, Faculty of Geography, Philipps-University of
Marburg, Deutschhausstraße 10, 35032 Marburg, Germany,
E-Mail: [email protected]
1
Cluster Growth and Institutional Barriers:
The Development of the Automobile Industry Cluster
in Shanghai, P.R. China
Abstract: Recent work has provided evidence that the establishment of new industry clusters can hardly be jumpstarted through policy initiatives alone. This does not, however, imply that the genesis of a new cluster cannot be planned at all. Especially in the context of a developing economy, it seems useful to re-investigate the relation between economic development, multinational firm strategies and state intervention in this respect. Drawing from the case of the automobile industry and its supplier system in Shanghai, in which German firms play an important role, we will provide empirical evidence how a new cluster evolves which is characterized by a focal, hierarchically-structured production system and how this is supported by the state in its various forms. We use a multi-dimensional approach to clusters which provides a more nuanced understanding of the evolution and growth of a cluster then that provided by earlier accounts. This approach allows us to distinguish the development of the Shanghai automobile industry cluster along its vertical, horizontal, external, institutional and power dimensions. We provide evidence that another dimension, i.e. ‘culture’, plays an important role, especially in its relation to issues of power and institutions. This is demonstrated in the case of German firms which tap into the Chinese innovation system. This system is characterized by particular business relations, institutions, norms and all sorts of social practices which are ‘new’ to the German firms and where they cannot draw from previous experience. We will demonstrate how this difference creates problems in establishing local production and supplier relations and how these can be overcome.
Keywords: clusters, institutions, power, ‘inter-cultural’ communication, boundary spanners, automobile industry, Shanghai
JEL classifications: F23, L13, L23, L62, P20
2
1. Introduction
In recent years, one of the hotbeds of research in sociology, economics, political science
and economic geography has been to analyze the establishment of clusters of interrelated
firms, their continued growth and the way how they secure reproduction. From a study of a
larger number of industrial clusters in different parts of the world, Bresnahan, Gambardella
and Saxenian (2001) found, for instance, that the reasons behind the establishment of a
new cluster are much different from those which support the growth of an existing cluster.
In their comparative study of ICT (information and communication technology) clusters,
they found that external effects, agglomeration factors and networking synergies were
primarily important in the later stages of growth. In contrast, outstanding entrepreneurial
activities, the willingness to take the risk of starting up new ventures and the ability to tap
into areas outside the established technologies and markets had been decisive for the
genesis of the clusters. The success of these entrepreneurial endeavors particularly
depended on the firms’ abilities to access major markets outside the cluster in their early
stages. Bresnahan, Gambardella and Saxenian (2001) emphasize from their research that
state policy did not have a substantial influence on the establishment of the clusters
studied.
Although we agree with this study on most accounts, we are somewhat skeptical about the
generality of its conclusions. It might be true that the success of clusters cannot be
planned in advance and that state policies are not that important in the case of ICT
industries, due to the particular nature and organization of this industry. But especially in
the case of the automobile industry, the situation is somewhat different. The automobile
industry is characterized by a hierarchical social division of labor, organized in tiers around
powerful car manufacturers. Although this focal production network is far from being a
simple top-down hierarchy, it is characterized by very strong ties and uneven power
relations. The car manufacturers are a primary force driving the organization of the
production system and its spatial manifestations. When they decide to go abroad, they
demand that their key suppliers also establish production facilities in the host country,
preferably close-by. Due to its organization around a focal company, the automobile
industry is highly responsive to cluster formation when entering new market regions. This
development is largely planned as opposed to those cases investigated by Bresnahan,
Gambardella and Saxenian (2001). State policies might not be the major force driving the
genesis of a new automobile industry cluster but it would be somewhat naive to ignore such
influences, especially in the context of a developing country, such as the People’s Republic
of China, which still has some characteristics of a planned economy.
3
In this study, we have chosen a cluster perspective with the intention of providing a better
understanding of the localized nature of international production networks.1 There are
several reasons which favor such an approach in the context of this study.2 Firstly, the
organization of automobile production in Shanghai around Volkswagen (VW) and General
Motors (GM) is highly localized. Secondly, as seen through a geographical lens, the
developing production cluster and producer-user relations in the automobile industry are
more persistent than, for instance, those in fashion-related consumer-goods industries,
such as textiles, footwear and apparel (e.g. Gereffi and Korzeniewicz 1990, Gereffi 1994,
1999). Thirdly, the approach is well suited to explain cluster formation in response to high
spatial transaction costs and difficulties to realize untraded interdependencies in an ‘inter-
cultural’ context.
This paper is organized as follows. In section 2, we develop a conceptual framework for the
analysis of clusters which is build upon a multi-dimensional approach. This section also
emphasizes missing links in the cluster literature, drawing on the importance of ‘culture’,
institutions and power in network formation. In section 3, we focus on the genesis of the
vertical and horizontal dimensions of the automobile industry cluster in Shanghai and the
1/. The arguments presented in this paper draw from research which was conducted in the Shanghai
automobile industry cluster. The primary goals of this research were to understand the successes and failures
of German firms which have established production facilities and supplier networks in Shanghai and to
elaborate on the role of culture and institutions in this process. Between 2001 and 2003, five research trips
were organized and more than 50 personal interviews conducted with representatives of automobile firms,
their German and Chinese suppliers, as well as formal institutions which support the growth of the industry.
Further, field study analysis was conducted in regular staff and strategy meetings and walking tours through
the productions facilities of those firms investigated. In addition, exploratory interviews were conducted at
the German sites of some of these firms.
2/. As Humphrey and Schmitz (2002) have pointed out, there are basically two different conceptions which
could be used as a basis to study the process how industries extend their production networks to other
countries, i.e. clusters and global commodity chains. Each of these approaches is associated with particular
problems in this context. The cluster concept (Porter 1998, 2000) explains economic success by drawing on
the internal structure of social relations between local or regional firms while neglecting extra-local linkages.
In contrast, the concept of global commodity chains (Gereffi 1994, 1999) emphasizes the advantages of
international production organization and governance structures but underestimates the territorial dimension
and the localized nature of production arrangements. In order to bring both perspectives closer together,
Dicken, Kelly, Olds and Yeung (2001) and Hendersen, Dicken, Hess, Coe and Yeung (2002) develop a concept
of global production networks which is spatially sensitive because it builds on actor-networks and social and
territorial embeddedness. Humphrey and Schmitz (2002) deal with this by applying a global value chain
perspective which provides an analysis of the localized effects of upgrading strategies.
4
role of governmental authorities in this process. Section 4 identifies precursors of a
developing trans-national community in Shanghai. In this section, we deal with the
consequences of ‘cultural differences’ between Chinese and German actors. Therein, we
emphasize the role of boundary spanners and the process of institution building. In the
final section, we summarize our findings and make some concluding remarks.
2. Cluster dimensions in an ‘inter-cultural’ context
In this paper, we use the term cluster to refer to a local or regional concentration of
industrial firms and their support infrastructure which are closely interrelated through
traded and untraded interdependencies (Maskell 2001, Bathelt and Taylor 2002). Clusters
should be analyzed along several dimensions; that is, their horizontal, vertical,
institutional, external and power dimensions. Through this, different configurations of
clusters can be identified according to their development stage and growth potential
(Porter 1998, 2000, Malmberg and Maskell 2002, Bathelt 2002).
Such a conceptualization may help to overcome the shortcomings and simplifications which
Martin and Sunley (2003) have identified in much of the cluster literature. In this paper, we
wish to emphasize another poorly developed link in the literature on clusters which
becomes particularly important in the context of international production arrangements.
The cluster literature tends to under-conceptualize issues of power and ‘culture’. This is, in
part, due to the fact that empirical studies focus on well-functioning clusters in particular
regions. In such a context, efficient communication processes between the cluster actors
which enable reproduction within a homogeneous ‘cultural’ and institutional environment
already exist. Such an environment does not exist though for automobile firms trying to
establish new production clusters outside their homebase, such as VW and GM in China.
Here, the existing ‘cultural’ and institutional frameworks which shape human interaction
differ substantially between home and host country. Automobile firms establishing
production in Shanghai have to bridge these differences, establish efficient communication
processes between agents with various ‘cultural’ backgrounds and adjust organizational
practices originating from the home country to those in the host country. We develop a
conceptual basis in the following subsections which particularly draws upon these ‘cultural’,
institutional and power dimensions of a cluster.
5
2.1 Cooperation and competition: the vertical and horizontal dimensions of a cluster
When discussing the advantages of being located within a cluster, most studies refer to
those firms of a value chain which are linked through supplier and customer relations. This
vertical dimension consists of firms with complementary products and competencies. They
benefit from intensive transactions within the cluster and form networks of traded
interdependencies. The idea behind this is that an existing agglomeration of specialized
producers gives rise to a substantial demand for specialized inputs. This creates an
incentive for suppliers and service firms to move close to these customers to supply the
growing regional market (Marshall 1920). The firms, in turn, benefit from low transportation
and transactions costs, as well as economies of scale, and thus gain a competitive
advantage (Scott 1988, Krugman 1991, 2000, Fujita, Krugman and Venables 1999). In part,
this explains why existing clusters tend to grow, giving rise to labor market specialization.
The reduction of transaction costs is, in fact, an important reasoning to understand why
automobile suppliers in Shanghai have a much stronger tendency to agglomerate than their
counterparts in Germany. VW demanded that or offered incentives to their suppliers to
locate within China. This was also due to the pressure of the Chinese government to fulfill
local-content regulations. Mostly, the firms established production facilities within the
larger Shanghai region (Depner 2003). The concentration in the region is extremely
important as it reduces the costs of communication, knowledge-transfers and adjustments
in products and processes.
Of course, the German firms in this cluster do not exclusively do business with each other or
rely on imports from their overseas operations. They have also begun to establish relations
with Chinese suppliers. This is, however, a difficult process as knowledge about Chinese
firms is quite limited. As one manager mentioned, new suppliers appear on their doorstep
on a daily basis and offer their services (Interview 3). In this respect, co-location within
the cluster serves as an important mechanism to access information about potential
suppliers. It also reduces the costs of supervising them.
The above discussion, however, shows that the advantages of clustering cannot be grasped
by concepts of cost and traded interdependencies alone. In emphasizing localized
capabilities (Maskell and Malmberg 1999a, 1999b) and untraded interdependencies (Storper
1995, 1997), recent studies have provided evidence that clusters offer manifold
opportunities for inter-firm communication and interactive learning within the same socio-
institutional and technological environment (Lundvall 1988, Gertler 1993, 1995). This
generates specialized information flows and supports innovation (see, also, Cooke and
Morgan 1998, Cooke 1999, Lawson 1999, Gordon and McCann 2000, Bathelt and Jentsch
2002).
6
In a similar way, advantages from a local or regional concentration of competing firms also
cannot be easily measured. This horizontal dimension of firms producing similar products is
often neglected in studies of clusters, although this dimension is sometimes key to
understanding why a cluster exists to begin with (Porter 1990, 1998, Maskell 2001,
Malmberg and Maskell 2002). Since these firms produce the same sort of products and have
similar competencies they have little reason to cooperate. Co-location provides the
opportunity to closely watch other firms and compare their economic performance with that
of others (Maskell 2001, Grabher 2001). This is possible because firms in a cluster basically
operate under the same conditions, sharing the same labor market, the same set of local
suppliers and essentially the same cost structure. This creates a competitiveness to
outperform others. It serves as a strong incentive for product differentiation, process
optimizing or cost reduction, depending on the information about the competitor’s
strategies, product decisions and technologies.
In the context of the Shanghai automobile industry cluster, competition between firms is
quite limited, at least with respect to foreign producers and suppliers. This is related to the
fact that the production system of the automobile industry is characterized by a hierarchical
governance structure. Inter-firm linkages are primarily oriented upwards or downwards the
value chain, according to the requirements of the focal agent. Although linkages exist
between suppliers in the same tier, most relationships are vertical in character. This is quite
different from a localized network in an ideal-type industrial district (e.g. Scott 1988),
where interrelated small- and medium-sized producers interact with one another in many
different ways, without a single dominant producer.
Co-location and face-to-face contacts within a cluster also give rise to additional
advantages related to the circulation of information and inspiration (Bathelt, Malmberg and
Maskell 2002). This creates a particular information and communication ecology or ‘buzz’
(Storper and Venables 2002) in a cluster. It is related to constant flows of information and
updates of this information, intended and unanticipated learning processes in organized
and accidental meetings, based on the same understanding, interpretative schemes and
technology attitudes within a particular value chain (Bathelt, Malmberg and Maskell 2002).
The importance of this ‘buzz’ is that actors do not have to search their environment or make
particular investments to get access to this information. They are automatically exposed to
news reports, gossiping, rumors and recommendations about technologies, markets and
strategies, by just ‘being there’ (Gertler 1995, Grabher 2002a).
In the Shanghai automobile industry, we would also expect that this ‘buzz’ plays an
important role, despite the fact that the horizontal dimension is still in an infant stage and
the communication ecology limited. Contacts with suppliers, customers and trade
7
organizations, executive meetings and social activities within the professional community
help to stimulate some sort of ‘buzz’ about governmental agencies, local suppliers and day-
to-day problems. This ‘buzz’ is even more important for the Chinese joint venture partners
who engage traditionally in personal relationships to obtain, for instance, access to
contracts or political decision makers. Such dyadic relationships or ‘guanxi’ are still quite
important in the Chinese society. They are open-ended in character and involve an ongoing
‘give and take’ of those actors involved.3 ‘Guanxi’ and the wider business networks which
develop from these dyadic relations, through the effects of recommendation and day-to-day
practices (‘guanxi wang’), are characterized by long-term social relations and continuous
communication between the network members. This stimulates fine-grained information
transfer. However, this Chinese ‘buzz’ does not always make its way to the foreign joint
venture partners. As will be shown in section 4, information exchange between joint
venture partners is often fairly problematic, indirect and incomplete.
2.2 Trans-local linkages and the external cluster dimension
It is clear that a cluster cannot fully unfold its growth potential if the firms exclusively rely
on internal markets and knowledge circulating through the local ‘buzz’. If the social
relations inside a cluster become too rigid and too exclusive, focussing on only a few local
actors, this could cause problems of lock-in, gullibility and blind confidence (Granovetter
1973, Kern 1996, Oinas 1997, Scott 1998, Maillat 1998). This phenomenon of being too
inward-looking, which has been described as ‘over-embeddedness’ (Uzzi 1997, Bathelt 2002,
Sofer and Schnell 2002), could cause firms to overlook technologies developed outside and
3/. These relationships, in combination with a strong hierarchical society structure, defined rules how a person
should act in certain situations, depending on the type of relation he/she had with another person. ‘Guanxi’
were never one-sided relationships but were always associated with obligations and returns on both sides (Hsu
and Saxenian 2000, Yeung 2000, Wang 2001, Schramm and Taube 2002). Each side would have to fulfill its
commitments in order not to lose ‘face’ (Hwang 1987). The latter would cause some sort of social isolation. In
such a society, which was based on clear rules and commitments according to the structure of social relations,
the law system and contracts did not play an important role. The importance of personal relationships and the
attached rules are especially problematic for foreign companies as they are used to relying on the authority of
contract and law, instead of more-or-less tacit, unwritten rules and regulations. Foreign firms which look for
Chinese partners often do not understand that their counterparts show little interest in the negotiation of
contracts. And, if they have agreed upon a contract with their Chinese business partners, they could easily
draw the wrong conclusion from this, assuming that a business relation would easily develop based on formal
agreements.
8
slow down innovation. Of course, this problem is always present in the automobile industry,
where most advanced suppliers rely on strong ties with few customers, organized in a
hierarchy of tiers. ‘Openness’ in this production system is limited but remains an important
issue for individual firms (Bathelt and Taylor 2002).
Many studies provide evidence that decisive triggers for innovation result from trans-local
partnerships with leading actors and firms from other regions (see Bathelt and Taylor 2002).
When firms consciously establish such ‘pipelines’ (Owen-Smith and Powell 2002) they can
get access to knowledge pools and markets outside the cluster. In this case local ‘buzz’ and
trans-local ‘pipelines’ can be mutually reinforcing. The more firms engage in ‘pipeline’
development, the more additional information and knowledge enters the local ‘buzz’, driving
product and process development. In turn, this knowledge enables firms to build up new
ties to external actors (Bathelt, Malmberg and Maskell 2002).4
An important pre-requisite for a firm to successfully establish ‘pipelines’ is the ability to
assimilate the information arriving through pipelines and apply it successfully. Cohen and
Levinthal (1990) have labeled this ability as being a firm’s ‘absorptive capacity’ (Malecki
2000). This ‘absorptive capacity’ depends on the firm’s direct interface with its local
environment and on the number and extent of its ‘pipelines’. It also depends on the way in
which information can be transferred across and within sub-units of the firm. The role of
internal gatekeepers and boundary spanners, which we will discuss further in the context of
relational power, becomes crucial for translating externally produced knowledge into a form
that can be internally understood (van den Bosch, Volberda and de Boer 1999, Giuliani
2002).
In the case of the automobile industry in Shanghai strong pipelines exist between the
German headquarters and their subsidiaries. VW’s engineering and development activities
are still concentrated at its headquarters in Germany, with the exception of those models
which have been designed for foreign markets, such as the VW Santana 2000 and VW Gol.
When the German headquarters of VW decides to make changes to a model which is not only
being produced solely in China but also in Europe, such as the VW Passat and VW Polo, the
respective suppliers in China get the exact specifications through their German headquarters
4/. In the case of VW, the establishment of production facilities in Shanghai can be viewed as an attempt to
develop a ‘growth periphery’ in the sense of Storper and Walker (1989) in order to obtain access to a new
consumer market and stay ahead of main competitors. The move into the Chinese market was a step to
actively develop ‘pipelines’ which also have a growth impact on the company’s home country operations.
Strategic competencies, of course, have largely remained in Germany.
9
(Interview 15). This rigid hierarchical structure is changing, however. As the suppliers
diversify their customer base in China, they develop a need to create developing and
engineering competencies within the country. Therefore, some suppliers have begun to set
up engineering departments in Shanghai (Interviews 33, 45, Case study 2).5
Exchange between operations in Germany and China in both directions also takes place, as
management and technical personnel are being sent from the headquarters to foreign
locations, who return after a few years. Most German executives in Shanghai have a three-
year contract, which can normally be extended. An obvious goal behind this strategy of
constantly employing executives, managers and technicians from Germany abroad is to
maintain control over the production activities in China. Another goal is to reintegrate
these managers in their respective German headquarters and, through this, improve
knowledge about the Chinese market, the ways how people interact and the rules according
to which processes are being organized in China. While some firms have a substantial
number of Germans who constantly work in their Shanghai locations, others have reduced
this number to a minimum. Often they replace German managers by Chinese experts who
have studied or lived in Germany and have acquired knowledge of both production contexts.
In such firms, German experts are usually appointed to work in the Shanghai facilities for
shorter time periods, often associated with the solution of a particular problem or the
introduction of technological and organizational innovations (Interviews 14, 23, 45, 47).
Further, Chinese managers and technicians are also being sent to Germany for training
purposes and to obtain work experience. This appears to be an efficient way to train them
and make them familiar with German routines, as has been pointed out by some German
executives (Interviews 27, 48). The advantage of training qualified Chinese employees
abroad is that they are not in their day-to-day working context where they would be
watched and evaluated by their co-employees.6
5/. An executive of a German-Chinese joint venture pointed out that this was absolutely decisive. Technicians
of a Chinese car manufacturer had come to him one day for help in the construction of a particular module. As
they were unable to explain the problem in technical terms, they simply presented the module to him
(Interview 35). To react to such a request for help requires engineering competencies be available.
6/. Another important link to firms in other countries results from the need to install and run specialized
machinery and equipment in the production process. Most of these machines are not produced in Shanghai or
other Chinese regions, as the machinery industry in China is still in an infant stage. German firms also prefer
to use the same type of machines at different locations and to cooperate with those machinery suppliers they
know already from Germany. Case study analysis showed that this tendency to rely on foreign machinery
producers is reinforced by negative experiences with Chinese machines which have to be repaired more
10
Over time, Shanghai operations of German automobile suppliers have been able to reduce
their dependence on their headquarters and make more decisions about day-to-day practices
and local market strategies on their own. This occurs as Germans are increasingly being
replaced by Chinese managers and producer-user linkages become more diversified, due to a
growing horizontal and vertical dimension of the cluster.
2.3 ‘Culture’ and institutions: underconceptualized links in the cluster literature
The ability to overcome barriers is most important in an international context where firms
operate across national borders and ‘cultural’ contexts. If firms are engaged in trans-local
linkages with partners from other parts of the world, they have to be able to understand
different institutional regimes in order to communicate and interact with their partners
(Owen-Smith and Powell 2002). This requires complex capabilities. The debate about the
‘cultural turn’ in economic geography has shown that ‘cultural’ contexts are an important
influence on economic action which cannot be neglected (Crang 1997, Sayer 1997, Boeckler
1999, Thrift 2000).
We are aware that it is quite difficult and problematic to define ‘culture’ (e.g. Lash and Urry
1994, Cavallardo 2001). There are many competing definitions and sometimes it appears
that the term ‘culture’ is used as a catchphrase to include almost every human practice. Our
intention is not to engage in this discussion. Rather, as Power (2002, p. 104f.) points out,
“it is useful to stress that in contemporary capitalism culture seems to have particular
characteristics that are constituted and bound into processes of definition and distinction
that are inherently woven into the fabric of relations of production, consumption, and
power that make up everyday life”. Of great importance for a representation of ‘culture’ are
the implicit and explicit norms, rules, convictions, moral codes and philosophies of life
which are accepted by or imposed on its members. These have developed through a history
of social relations and are shaped, produced and reproduced in everyday’s practices of
human action and interaction (Lash and Urry 1994, Sayer 1997, Cavallardo 2001, Hösle
2002). Language, arts, symbols, artifacts, etc., help create meaning and ‘cultural identity’,
which are different from that in other cultures. Through this, it is possible to distinguish
‘insiders’ from ‘outsiders’.
frequently. In many instances, Chinese machines are used for simple activities only. Often, they are simply
imitations of European machines (Case studies 2, 3, 4).
11
A ‘cultural’ context creates an institutional framework for ongoing human interaction. At
the same time, existing rules and norms are constantly being reshaped and new ones
created through interaction. Therefore, ‘culture’ cannot be reduced to a fixed number of
characteristics. It is a phenomenon in a state of flux. In this paper, we are interested in
which way ‘culture’ has an impact on human action and interaction in economic context. 7
Gertler (2001) points out that systematic influences of institutions, especially between
different national environments, prevent the diffusion of universal operational standards or
a single ‘best practice’ (see, also, Boyer 1998). As he states, “[t]he nation-state (whether
‘home’ or ‘host’) is still a primary source of influence over industrial practices” (Gertler
2001: 16). It still seems to be a powerful umbrella to produce not only language, symbols,
artifacts and meanings associated with a ‘culture’, it also stimulates particular ways of
economic action and interaction through formal and informal institutions. The nation state
further shapes practices of learning (Tracey, Clark and Lawton Smith 2002). One way how
these structures and practices are being reproduced is by the creation of social identities
and ‘othering’ (Berndt 2001, Cavallardo 2001). It is quite possible that a lack of
understanding of different institutional contexts may block communication. In this case,
firms would continue ‘speaking different languages’.
Foreign automobile firms, establishing a new production cluster in Shanghai, have to face
this problem of bridging ‘cultural differences’ in a number of cutting points. First, they have
to develop viable communication structures with their Chinese joint venture partners to
establish coherent management systems and gain access to external information about state
regulations, market developments and labor market characteristics on the national and
regional level. Second, it is necessary to train Chinese workers to develop work routines and
enable interaction between these workers and German managers and technicians. Third,
routines will also have to be established to integrate Chinese suppliers into the production
system. Fourth, if the German firms intend to create ‘pipelines’ to other Chinese regions
they have to acquire information about potential partners in order to evaluate them.
7/. This is not to say that ‘culture’ is a homogenizing phenomenon. In relational economic geography (see
Bathelt and Glückler 2003), ‘culture’ can never be a phenomenon which fully determines human action and
interaction. It does create, however, patterns which are more or less accepted and have, over time,
established structures and habits which make some actions appear more likely than others (Sayer 1997, Hösle
2002). There is, nonetheless, always a fundamental openness as in which way an individual may decide in a
particular situation. When we talk about views and habits of Chinese and German agents in this paper we do
not imply that such behavior can be projected into simple categories.
12
The successful establishment of ‘inter-cultural’ communication in the Shanghai automobile
industry cluster therefore requires the development of a shared institutional context which
enables interaction and knowledge creation. It also requires the establishment of methods
and routines to control production and pass on orders and messages effectively within the
production system. The joint institutional framework enables specialized users and
producers to discuss and solve particular problems (Hodgson 1988, North 1990). Such an
institutional framework does not, however, exist spontaneously. It is created through social
practices in day-to-day interactions between the actors of a cluster. In these interactions,
joint problem-solving and experimentation lead to preliminary fixes which must be robust in
order to survive the next series of interactions. These fixes are constantly updated or
adjusted to new goals in the innovation process (Storper 1997, McKelvey 1997).
Co-presence and co-location within a cluster serve as a powerful means to participate in the
process of creating institutions. It enables firms to make sense of the local ‘buzz’ and
distinguish between valuable and unimportant information. This happens as firms develop a
similar language, attitude toward technology and interpretative schemes (Lawson and
Lorenz 1999). Over time, ongoing interaction may give rise to what could be phrased as a
high-trust environment within the cluster (Maskell, Eskelinen, Hannibalsson, Malmberg and
Vatne 1998).
This role of institutions becomes much less straightforward in a multinational context, such
as the automobile industry cluster in Shanghai, where different rules of the game exist
which are not easy to grasp (Lundvall and Maskell 2000, Bathelt and Depner 2003). German
firms are used to operate within an existing institutional framework, which has been shaped
in their home country over a long time period. The institutions they are faced with in the
Shanghai region result from a different national context. Some of the institutions may be
supportive to the production system of the German investors. Many others, however, are
different and not compatible. Another problem arises since the institutional conditions in
China are not fully known to the German firms in advance. This creates a great deal of
uncertainty in transactions. In order to create a production system similar to the one in
Germany, institutions will have to be imported from Germany to provide the basis for
interactions within and between firms. In addition, a new set of institutions will have to be
created to enable effective communication between the Chinese and German management
and workforce in joint ventures.
The experiences of firms within a particular national context provide evidence that the
process of institution building can be triggered or strongly supported by the establishment
of ‘communities of practice’ (Brown and Duguid 1991, Wenger 1998). Wenger and Snyder
(2000: 139) define such communities as “groups of people informally bound together by
13
shared experience and passion for a joint enterprise ...”. ‘Communities of practice’ consist of
agents which are bound together through day-to-day interaction of various kinds, such as
e-mail networks, social events and regular feedback meetings. Members of these
communities discuss their work to bridge the gap between theoretical and practical
knowledge (Brown and Duguid 2000). This is beneficial in processes of problem-solving,
strategy and business development and the diffusion of ‘best practice’. Specialists
voluntarily join these communities, adding passion, commitment and an identification with
the group’s competence. This holds the communities together (Wenger and Snyder 2000).
‘Communities of practice’ can develop within a firm but may also span a single organization
and include other firms (Gertler 2001, Grabher 2002b). Shared experiences in solving the
same sort of problems, based on the same basic understanding and technological paradigm,
support the development of mutual engagement, joint enterprises, shared repertoire and
negotiation of meaning (Wenger 1998, Coe and Bunnell 2003). Clusters can become
important catalysts for the formation of such communities. In this case, they develop into
local frames to understand the meaning and significance of local ‘buzz’ which in turn serves
to stimulate the generation of more local ‘buzz’ and its rapid diffusion.
It would not be easy, however, to establish ‘communities of practice’ in the automobile
industry cluster in Shanghai. Individual firms send only few specialists to their foreign
branches, once they start regular production activities. As communication with foreign
workers is limited because of language problems and ‘cultural distance’ (Gertler 1997), the
potential to develop communities within firms is, at this point, fairly low. At the same time,
communication and interaction between technical experts of different firms is also limited.
Only rarely do German specialists of different automobile suppliers get together to discuss
problems and exchange solutions, partially because the competencies are too diversified to
engage in an inter-firm community. The German branches also focus on few stages and
selected functions in the overall production process. The most likely scenario for community
establishment, aside from the already existing ‘organized’ community of Chinese managers
and technicians, is between the foreign executives of different firms. At this level,
managers can share their experiences in dealing with their Chinese joint venture partners,
the difficulties they face when designing labor processes for Chinese workers and the
strategies they use to get information about the quality of Chinese suppliers.
2.4 Relational power and the role of ‘inter-cultural’ boundary spanners
A cluster does not automatically encompass a coherent group of firms which cooperate
harmoniously with one another to achieve a common set of goals. A cluster’s horizontal and
14
vertical dimensions are shaped by existing power relations and asymmetries, which impact
the agents’ ability to react to changes in their regulatory environment and external markets
(Taylor 2000; Bathelt and Taylor 2002). In actor-network theory, those actors who are
viewed to have power are able to build networks and develop them further by enrolling
others actors (e.g. Murdoch 1995, Smith 2003). This is the sort of power referred to by
Latour (1986) as the ‘powers of association’ or as ‘power as relationships’ (Allen 1997,
Taylor 2000).
In a cluster context, it is necessary to consider the whole structure of relations within the
actor-network. These social relations are constantly being produced and reproduced through
ongoing communication between the actors. This is not a simple diffusion of information
from one end of the cluster to the other. Rather, this should be viewed as a translation
process where messages are being transferred to other actors through social relations who
evaluate them according to their goals. During this transfer, the messages are constantly
being interpreted and reinterpreted by those actors involved, giving each actor the
opportunity to apply changes (Latour 1986). The power of a cluster can thus be defined as
the potential to enroll actors in joint enterprises. Since the coherence of a cluster and its
ability to work are dependent on day-to-day interactions between its actors and firms,
distance and visibility are of great importance. They enable ongoing interaction with others
and to exercise control over their activities.8
According to this relational view of power, network builders, who are able to enroll others
into networks, are particularly important in the process of network activation and
formation. In an ‘inter-cultural’ context, the role of boundary spanners also becomes
important as they have the potential to communicate between the people involved and
provide an understanding of heterogeneous habits and attitudes. Coe and Bunnell (2003)
emphasize the importance of transnational communities in translating news, claims and
8 Of course, it is difficult to establish coherence within a cluster through social relations alone. Material and
non-material resources, such as non-human artifacts (e.g. particular technologies, symbols), tools (e.g.
manuals, reports) and accepted rules, enable human actors to engage in social relations and keep them going
(Murdoch 1995, Dicken, Kelly, Olds and Yeung 2001). They are the glue of social relations (Latour 1986). In a
cluster context, particular process and communication technologies serve to stabilize interactions between the
actors of a cluster, as they have similar day-to-day experiences and develop the same understanding.
Especially, in an ‘inter-cultural’ context shared technologies and existing norms and rules are important to
support the interaction between people from different ‘cultural’ contexts in achieving common goals. Such
material and non-material resources also shape the course of action. Without them, messages could, for
instance, be easily misinterpreted by other actors and technologies used differently.
15
problems back and forth between the different nationalities and parties in a cluster (see,
also, Hsu and Saxenian 2000, Smith 2003). These are people who have lived and worked in
different ‘cultural’ contexts and, from this, are able to understand the different expectations
and patterns of behavior and clarify them between actors. They are not to be confused with
traditional immigrants as they develop professional expertise and maintain active contacts
with both ‘cultures’ they are related to. Saxenian and Hsu (2001: 915f.) conclude from their
case study of US-Taiwanese linkages in high-technology industries that “[a] transnational
technical community allows distant producers to specialize and collaborate to upgrade their
capabilities, particularly when the collaborations require close communication and joint
problem-solving. The trust and local knowledge that exist within technical communities,
even those that span continents, provide a competitive advantage ...”.
In the context of the Shanghai automobile industry cluster, such boundary spanners are
extremely valuable in communicating between Chinese and German managers and workers.
Chinese boundary spanners can also help Germans to get access to the Chinese political and
law system. This can be very beneficial when making decisions about strategies and
investments in the future. However, this potential is not easy to mobilize, especially when
the German management does not view Chinese joint venture partners as experts.
Relationships with the Chinese management in joint ventures are often seen as a burden
and remain limited. At the same time, the Chinese management does not do much to
intensify these relationships either. In short, the argument is that a lot of inefficiencies
and problems could be avoided if knowledgeable ‘inter-cultural’ boundary spanners would be
employed by German or other foreign firms entering the Chinese market. Within joint
venture firms, this could help stimulate the creation of shared systems of meaning and
allow access to ‘inter-cultural’ knowledge pools.
The conceptual framework developed in section 2, which draws particularly upon the
‘cultural’, institutional and power dimensions of a cluster, serves as a basis to investigate
the establishment of a new automobile industry cluster in Shanghai. In the sections 3 and
4, we investigate how this cluster has been initiated and shaped by political influences and
analyze the cluster along its dimensions.
3. Genesis and growth of the automobile industry cluster in Shanghai
When China undertook its first steps to open the economy by designating ‘special economic
zones’ and ‘open cities’ and directing foreign investment to these areas Shanghai was not
included. The city was regarded as the central industrial core in the country which should
not be part of economic experiments. In the early 1990s, the government regretted this
16
decision and opened the city for foreign investors. Special support was granted to those
firms which established a business in the ‘Pudong New Area’, an area which previously had
been under agricultural use. The automotive industry was selected as one of a total of five
‘pillar industries’ to shape the future development of the city. After the financial crisis in
Asia, the municipal government strengthened its support of the manufacturing industry and
postponed plans to focus particularly on the financial and trade sector (Zeng 2000).
3.1 The role of SAIC as a nucleus of the cluster
The Shanghai Automotive Industrial Corporation (SAIC) is the dominant Chinese player in
the Shanghai automotive industry (Figure 1). As the group is largely state-owned, with the
majority of its shares being controlled by the city of Shanghai, SAIC is closely interlinked
with and supported by the city’s policy makers. It is more, however, than a remnant of the
old era of state monopolies. The SAIC group consists of different companies and
establishments which manufacture cars, trucks, busses and motorcycles, as well as parts and
equipment. By the end of 2001, SAIC had established 55 joint ventures with other
automobile and component manufacturers and employed almost 62,000 people (SAIC 2002).
As a leading car manufacturer in China, the group strives to become competitive at an
international level. The operations of SAIC are not limited to Shanghai (Figure 1). Among
others, the group holds major shares of the Liuzhou Wuling Automotive Company and
Jiangsu Yizheng Automotive Company, as well as shares of Anhui Chery (SAIC 2002). In
2002, SAIC acquired a 10% stake in GM’s South Korean operations (China Daily 2002).
Figure 1: Financial control over other companies as exercised by the Shanghai Automotive
Industrial Corporation, 2002 (Sources: SAIC 2002, survey results).
17
The Shanghai automobile cluster developed thanks to massive governmental support at
different levels. In the beginning, the central government’s initiative was key to attract VW
to China. In order to upgrade the national automobile industry to international standards
and to avoid having to import automobiles at a large scale, the central government started
negotiations with Volkswagen (VW) from Germany already in 1978. These resulted in a joint
venture agreement between the SAIC, Bank of China, China National Automotive Industrial
Corporation (the parent organization of the Chinese automobile industry) and VW. The joint
venture, which was called Shanghai Volkswagen (SVW), started to produce the ‘VW Santana’
in 1985.
In the first years of production, SVW still imported most parts and components needed from
overseas, a large part of which from Germany. At that time, there were basically no firms
within the region who could have supplied the parts needed. For VW’s German suppliers the
number of VW Santanas produced in Shanghai in the 1980s was too small to establish
production facilities in China. In 1990, SVW still assembled less than 20,000 cars per year.
In the late 1980s, the Chinese government threatened to enforce a production limit upon
SVW if the firm would not increase its local content in production. The idea behind this
policy was to stimulate broad manufacturing competencies within the region and integrate
Chinese suppliers, instead of importing key supplies from abroad.
In order to fulfill the regulations, SVW had to enroll Chinese parts producers into the
production. Zhu Rongji, who became the mayor of Shanghai in the 1990s, was especially
interested in supporting the growth of the automobile industry and, at that time, mediated
between the central government and SVW (Harwit 1992). Local suppliers were tied together
in the ‘Shanghai Santana Local Content Cooperative’. The municipality supported this
initiative by increasing the prices of cars by 16% and by providing low-interest loans to the
local suppliers (Lee, Chen and Fujimoto 1996). Additionally, the policy was established that
each taxi in the city had to be a VW Santana (Sit and Liu 2000). In 1991, many of the parts
producers were integrated into the SAIC group. In addition to the enrolment of local
suppliers VW demanded that some of its German first-tier suppliers establish production
facilities in China, preferably within the region.
As a consequence of this, the vertical cluster dimension has developed quite rapidly. The
growth of the cluster later on was very much supported by different policies of the
municipal authorities, such as infrastructure development, labor market and industrial
policies (e.g. Harwit 1992, Zeng 2000). The resulting supplier network is one of the most
advanced in China (Yang 1995).
Still, this was not enough for SVW to establish a hierarchical supplier network, similar to
the way it operated in Germany. Since the 1980s, there had been a strong tendency in the
18
international automobile industry to develop hierarchical supplier networks and pass
development, manufacturing and assembly responsibilities of important modules to the first
tier of suppliers. This reorganization enabled firms to reduce the number of direct suppliers
substantially (Gaebe 1993, Schamp 1995). Due to the strategy of automobile producers to
limit their suppliers to no more than two for each module (Veloso and Kumar 2002), the
first-tier suppliers were increasingly required to follow their clients to other countries when
these established new production facilities. This trend also had consequences for the
Shanghai automobile industry. In 1993, there were only nine foreign suppliers with local
operations in the Shanghai region (Yang 1995). But this changed quickly in the following
years. Especially, first-tier suppliers from Germany began to set up facilities in Shanghai.
3.2 The SAIC joint ventures: triggering vertical and horizontal growth
The majority of German suppliers chose to enter the Chinese market by establishing a joint
venture with one of the SAIC firms or branches (Figure 2). According to one representative,
this was not the free will of the German suppliers in all cases (Interview 26). At least in
some instances, SAIC demanded that those firms entering the supplier network establish
German-Chinese joint ventures and set up their facilities in Shanghai.9 In 2002, SVW had
371 direct suppliers in China. Compared to its European operations, this is a much larger
share of direct supplier relations (Interview 26). As has also been shown for other
automotive supplier networks in China (Veloso and Kumar 2002), this indicates that the
restructuring of supplier relations into a multi-tier system is not as advanced in China as it
is in Europe.
Joint ventures between German suppliers and SAIC seem to offer advantages to both sides.
SAIC benefits from the technology transfer. The German firms are able to use existing
production facilities, acquire parts and components through already established channels
and have access to all partners of SAIC, including the automobile producers. As will be
shown later on, day-to-day interactions can, however, be very complicated and inefficient.
As a result, a few German firms went a different route by setting up wholly-owned foreign
enterprises (WOFOEs) or establishing a joint venture with non-SAIC firms. This is not done
without risk, however. In an institutional context, which relies heavily on close personal
9 SAIC’s strategy in this process obviously was to integrate as many suppliers as possible into its own group
networks in order to develop broad competencies in the production of automobiles. Eventually, this would
enable the group to produce cars on her own. With the production of the ‘SAIC-Chery’, the group has recently
shown that it already has the potential to develop and produce cars without the aid of foreign partners.
19
relations to get access to resources, one can easily get locked out of the system. One
manager of a Sino-German joint venture, established without the participation of SAIC, was
angry because his firm ran into difficulties, as it had to undergo unusually hard,
uncooperative audits which the manager referred to as an ‘inquisition’ (Interview 17). Other
managers in similar situations mentioned that they were under much greater pressure to cut
costs than their SAIC counterparts (Interviews 18, 31).
Figure 2: Idealized structure of the automobile industry cluster in Shanghai and its external
linkages.
Until 2001, most German suppliers were able to make profits through their Shanghai
operations despite the relatively small number of units they produced. Some of the SAIC-
German joint ventures reached their break-even point already during the first year of
production. Since then, however, competition has increased and price control been lifted,
20
creating pressure on the automobile producers to reduce car prices.10 One interviewee
mentioned that, as a consequence, SVW demanded its suppliers to reduce cost by 10-40% in
2002 (Interview 35). This price pressure serves as an incentive for the suppliers to localize
their parts production instead of importing parts and components at high costs.
At the same time, the establishment of parts production within the region is also time-
consuming and requires substantial investments. Often there is no alternative, however.
Although there are approximately 5,500 suppliers in China (Roland Berger Consultants
2001), it is very difficult to find high-quality, technologically-sophisticated firms which the
German suppliers can rely on. Several managers blamed Chinese firms for claiming that they
are able to produce parts according to the exact specifications even if they were not able to
meet the required standards (Interviews 1, 3, 14, 15). An important task for German firms is
therefore to find reliable Chinese suppliers. Particularly complex and sophisticated parts,
such as electronic components and air bags, are typically still being imported.11 Many
interviewees also criticized the low quality of parts produced by Chinese suppliers
(Interviews 22, 30, 42, 52). In order to increase quality standards, some have begun
irregular visits at their suppliers’ plants to check up on quality. This requires, of course, that
the Chinese suppliers are ‘visible’ and located close-by.
Despite these problems in producer-user relations, the policy to increase local-content
production has generated a growing vertical dimension of the automobile industry cluster.
Supposedly, this industry infrastructure which had been created was one of the reasons why
GM also decided to set up production facilities in Shanghai in 1997, as the firm was able to
benefit from an already existing network of suppliers.
10/. The year 2001 was a milestone for the Chinese automobile industry, as the country joined the WTO. As a
prerequisite for this, customs-duties for imported goods had to be reduced and import quotas increased, thus
allowing for increased competition from the world market.
11/. Another interviewee explained that his firm did not intend to expand its local supplier linkages further as
they had already reached a certain level. He was fed up with the process of establishing a supplier relationship
with a Chinese firm, as it requires too much attention and manpower. There are too many discussions involved
in the course of introducing and maintaining producer-user relations. Having engaged in a number of relations
with Chinese suppliers, the firm noticed that these same suppliers had also begun to produce parts for its
competitors (Interview 46). Another manager pointed out that his firm does not even try to integrate Chinese
parts producers and, instead, continues to rely on imports (Interview 17). Even a Chinese executive
complained about Chinese suppliers: “They don’t have the management capabilities, they have no capital and
no technologies. I don’t want to work together with these companies any more.” (Interview 34).
21
Prior to 1997, SVW was the only car manufacturer in Shanghai and had a network of
exclusive suppliers. Back then, GM decided to set up manufacturing operations in Pudong in
a joint venture with SAIC, i.e. Shanghai General Motors (SGM). Although SVW is still the
major customer of the German automobile suppliers, increasing shares of their activities are
now devoted to the production of parts for SGM and other car manufacturers. For a long
time period, SVW had only introduced one model, i.e. the VW Santana, into the Chinese
market and was a typical transplant in a developing country (Lee, Chen and Fujimoto 1996).
To reduce its dependence on VW and to put pressure that it changes its product policy and
increases technology transfer, SAIC decided to engage in the joint venture with GM. Since
its establishment, SGM has grown into one of the largest car manufacturers in China.
Between 2001 and 2002, GM increased its market share in China substantially from 2.7 to
7.7%, mainly due to its Shanghai operations (Vwd: Asien 2003b).12
GM’s entry into the Chinese market has initiated a number of changes in the product
strategy of SVW. The firm now produces several different models in Shanghai, i.e. the
‘Santana’, ‘Santana 2000’ (a model which was customized for the Brazilian and Chinese
market), ‘Passat’ (since 2000), ‘Polo’ (since 2001) and most recently the ‘Gol’ (since 2002),
which was originally designed for Brazil. One interviewee pointed out that VW now pushes
forward to the establishment of broad engineering competencies at SVW, to be able to
adapt cars to the specificities of the Chinese market (Interview 51).
SAIC is also involved in the national joint venture SAIC-Chery Automobile Company (Figure
1). Although SAIC-Chery’s operations are located a fair distance away from Shanghai in the
Chinese province Anhui, this new manufacturer buys parts and components from SVW’s
German-Chinese suppliers. The involvement of SAIC in this joint venture has not only
enabled the Shanghai suppliers to develop new linkages, they also deliver parts and
components which are similar to those produced for SVW. This was a move, however, that
VW Germany did not agree with. VW regards the parts and components which have been
developed by the suppliers for SVW as its intellectual property. According to one executive,
headquarters of the suppliers in Germany were thus asked to stop deliveries for SAIC-Chery.
Otherwise they would suffer negative consequences, also in their German relationships
(Interview 8). The suppliers in Shanghai did not agree with this policy as they benefit, at
this point, from a growing horizontal dimension in the automobile industry cluster.
12/. To increase its production capacity for the production of its model ‘Sail’, GM and SAIC have acquired
another firm in Shanghai with large production facilities. In the new plant, they plan to assemble 100,000
vehicles per year (Vwd: Asien 2003a).
22
Nonetheless, they had to follow VW’s decisions made in Germany in order to avoid conflict.
As a result, SAIC-Chery had to stop its operations because its warehouses with input
materials were empty.
At present, SAIC’s dominant position and power in the Shanghai automobile industry
remains unchallenged. As the group is closely tied to local authorities, banks and China’s
federal government in various ways, it has the power to exercise influence on the course of
the Shanghai automobile industry cluster. SAIC has benefited from the competition between
SVW and SGM, as it has absorbed much of the technologies imported and developed
manufacturing competencies. The SAIC-Chery venture has exemplified this quite clearly. To
be involved in several joint ventures also enables the firm to play one partner off against
another one and shift its resources accordingly. This does not exactly provide grounds for
the establishment of trust between the joint venture partners. Skepticism also increases
because SAIC organizes meetings in which Chinese managers from different operations and
joint ventures participate and exchange experiences. As they move from one joint venture
to another through the course of their career, they develop personal ties with many other
SAIC managers helping them to create something like a ‘community of practice’. Such
communities are not in the interest of the German joint venture partners, as this might
create stronger information flows between than within SAIC joint ventures.
4. Difficulties of establishing cooperation between Chinese and German
actors: issues of ‘culture’ and power
Having analyzed the development of the vertical and horizontal dimensions of the Shanghai
automobile industry cluster, we will now focus on the effects of ‘culture’, institutions and
power. In this section, we investigate how formal institutions are being imported to and
new institutions created within the Shanghai automobile industry. Further, we discuss the
intra-firm problems which arise from different norms, habits and expectations between the
German and Chinese management and workforce of joint ventures. We argue that boundary
spanners can become quite important to overcome such problems and mediate between
different positions in an ‘inter-cultural’ context.
4.1 Crossing the ‘cultural barrier’: Different tastes, expectations and behavior
The transfer of production arrangements and industrial practices to another national
institutional context requires adaptation to those social and economic structures of the
host country (Boyer 1998). It also requires that German firms adjust their products, at least
23
to some extent, to the customers’ tastes and behaviors to be successful in the new market.
SVW, for instance, had to make its Passat model longer in order to fit the taste of Chinese
consumers.13 Compared to product adjustments however, adapting to a different way of
doing business in another institutional context is even more difficult and can easily result
in failure. From our investigation, it seems that many firms underestimate the challenges
they face when they extend their business to another political, cultural and economic
setting. In a number of the joint ventures studied, for instance, either the German or the
Chinese part of the management had to be exchanged because of heavy communication and
interaction problems. German firms in China seem to be focussed on their technological
capabilities and the superiority of their products. In following their ‘missions’, they tend to
believe they can transfer their rationality and bureaucracy efficiently to the Chinese context
and underestimate the challenges awaiting them (Hoon-Halbauer 1999).
One important aspect, which cannot be ignored, is that the Communist Party (CP) still has a
strong influence on large firms (see, also, Hoon-Halbauer 1999). Larger joint ventures, for
instance, all have an extra CP office. German executives mentioned that it is virtually
impossible to change the internal organization of a joint venture and shift personnel from
one position to another if the CP does not agree with these changes. It seems also difficult
to fire a worker for whatever reason if he/she is a member of the CP (Case studies 2, 4).
When CP meetings are announced, they have a high priority in the Chinese workforce. If
this conflicts with other meetings, even those that have been scheduled much earlier, they
typically must be cancelled in favor of the CP meeting. Influenced by the CP, some Chinese
managers view a joint venture not just in terms of its business goals, but also see it as part
of a larger picture supporting the growth of the Chinese economy overall. As Zhu, Speece
and So (1998: 25-26) stated, “[m]anagers build careers by getting the foreigners to put in
money, technology and training to help build a Chinese company. … There is little reward
for helping make profits which will partly go out of China to the foreign partner.”
Another important aspect of the Chinese context is that communication within and between
firms is often based on personal relationships. Actors engage in the formation of dyadic
networks which are based on joint interest, interdependence, reciprocity, trust and open-
endedness. Such relationships are quite stable and long-term in character and are
13/. Another instance which created the need for customization were the complaints of Shanghai taxi drivers
that their VW Santanas’ horns and brakes would break too easily. As it turned out, failures were due to the
extreme stop-and-go traffic in Shanghai and the turbulent style of driving of the taxi drivers, using their horns
much more often than in VW’s home market (Interview 1).
24
associated with particular obligations and benefits on each side (Hsu and Saxenian 2000,
Yeung 2000, Wang 2001). Engagement in these ‘guanxi’ is shown through regular contacts
and reciprocal confirmations between those parties involved. The consequence of this
reliance on ‘guanxi’ in the social organization of production is that Chinese managers invest
a lot more time into the maintenance of these relationships and spend less time for
strategic planning and administrative duties than do their European and American
counterparts. ‘Guanxi’ obligations and returns can also be passed on to third parties (‘guanxi
wang’), if a Chinese manager who has established ‘guanxi’ with two formerly separate
parties wants to bring these together to form a larger network.14
4.2 Import and creation of institutions
Despite the attempts of the Shanghai municipal authorities to establish a business climate
which encourages foreign direct investment, the institutional context of China provides a
substantial barrier for those German firms who have no experience of operating in this
context. A lack of knowledge regarding the political, cultural and societal settings causes
problems when dealing with Chinese authorities, customers, employees and the like.
In order to provide support for German firms investing in Shanghai, the ‘German Chamber of
Industry and Trade’ and ‘German Chamber of Foreign Trade’ have jointly established an
organization, i.e. the ‘Delegation of German Industry and Commerce in Shanghai’ (GIC). The
experts of the organization give advice on how to set up production facilities, how to get
approvals from Chinese authorities and how to find appropriate Chinese suppliers (Interview
13). GIC also organizes cultural events and regular meetings for German technicians and
managers working in Shanghai in which they get information about new governmental
regulations and market trends and have opportunities to exchange information about
problems and experiences. There are even regular meetings for executives of automobile
firms and their suppliers. Such meetings are important in serving to provide social contacts
from within the automobile industry cluster in Shanghai which are otherwise not very
14 The Chinese executive of one firm which relies mostly on local Chinese suppliers stated that a German could
not do this job very well because he/she would not know how to establish and maintain personal relationships
with managers of the suppliers. He described that he had regular personal contact with these managers and
that these relationships went beyond the pure economic (Interview 29). This statement may, however,
underestimate the possibility of foreign individuals to obtain access to Chinese networks. In some cases,
German managers mentioned that they had themselves established close ties with their employees, customers
or suppliers, which helped them to improve interaction (Interviews 5, 16, 20, 34, 36, Case study 3).
25
common. Over time, this could lead to the establishment of something similar to
‘communities of practice’ (Brown and Duguid 1991, Wenger 1991), although this still has a
long way to go. In several interviews conducted, executives mentioned that they were not
overly interested in having intensive contact with other Germans outside their firm. They
work extremely long hours and, thus, do not have the time nor interest for additional
meetings (Interviews 16, 22, 27).
One of the problems of the Shanghai automobile industry cluster is that the local labor
market has not been able to satisfy the demand for specialized workers which has expanded
rapidly in conjunction with the growth of the industry. Firms seemingly have difficulties
finding suitable personnel and compete with one another for specially trained workers.15
Because of the shortages on the labor market, German firms and organizations have begun
to improve the vocational and professional training system in Shanghai. The ‘Hans Seidel
Stiftung’ (a political foundation which is based in Bavaria), for instance, opened a
vocational training center in which skilled workers of more than 100 firms are being
trained. At Tongji University, a Chinese-German university college was founded in 1998
where Chinese students can obtain specialized skills and receive different degrees required
by the industry. German universities, such as those in Bochum and Munich, participate in
professional training programs in sending teaching staff for temporary training. German
automobile firms support such initiatives. SVW and its German supplier Bosch, for instance,
have provided funds for additional chairs at Tongji University.
In order to ensure high quality throughout the local production chain, firms have enforced
quality control systems onto their suppliers which were developed within the German
automobile industry. One of the largest German organizations for the control of technical
and management systems in industrial firms, i.e. the ‘TÜV Rheinland-Brandenburg’, has, for
instance, established several branches in China. According to an executive of this
organization, the vast majority of first- and second-tier suppliers in Shanghai are already
certified according to the standards VDA 6.1 and QS 9000 of the German and American
automobile industry associations, respectively (Interview 39). Automobile producers
demand these certificates from their suppliers, although they also conduct their own
supplier audits as well. As Chinese organizations have also been founded which grant similar
certificates this becomes problematic. These certificates do not seem to have the same
value, however, as those granted by the German organization. Some German managers
15 An executive reported that the Chinese workforce is quite mobile. Some engineers would even leave their
workplace if a competitor offers a RMB 200 increase (US-$ 25) in their monthly salary (Interview 1).
26
complained about these certificates because the auditors seemingly do not exercise enough
care (Interviews 16, 19, Case studies 2, 4).16
4.3 Intra-firm tensions and potential communication failure
Theoretically, international joint ventures have a great potential to overcome ‘inter-cultural’
barriers and stimulate knowledge creation and interactive learning. In reality, this is quite a
difficult venture as it first requires that a common set of rules and routines be established
and supported by all parties. Therefore it seems arduous to establish efficient management
systems between the German and Chinese partners in joint ventures. Such a system would
require to introduce routines and practices which are accepted by both sides and enable
effective communication between the partners (Boyer 1998). According to Hoon-Halbauer
(1999), Chinese executives in joint ventures often regard the transfer of management know-
how as a change in their traditional ‘organizational culture’. This can become a major root
of conflict within a joint venture. On the other hand, executives at the German
headquarters can also be quite skeptical and distrustful of their Chinese counterparts which
provides another barrier for communication. According to Kiefer (1998), the performance of
SVW would have been better if the German management had attempted to integrate the
Chinese ‘guanxi’ system into their organizational structure, instead of pushing a hierarchical
Taylorist division of labor. According to him, things went better when Germans did not try
to intervene.
The circulation and diffusion of information provides a good example of differences which
exist between the German and Chinese organizational routines. In German firms, it is
normal to have regular business meetings in which managers are expected to scrutinize the
performance of their divisions and make suggestions for decisions regarding future
activities. This seems unusual in the Chinese context. In the meetings in which we
participated, decisions were hardly ever made. The Chinese managers did not seem to listen
carefully to what was being said (Case studies 2, 4). In fact, they used their mobile phones,
smoked cigarettes, talked to their neighbors and even periodically left the conference room
during meetings. We view this as a clear indication for miscommunication, as the Chinese
managers obviously did not regard these meetings as the ‘right’ setting to make decisions.
16 A Chinese manager at the other hand expressed his anger about the German certificates as they do not
entirely fit the realities of a Chinese production environment. He regarded these certificates as a tool of VW to
make market entry more difficult for Chinese suppliers. According to him, the dynamics of the industry would
make it extremely difficult to maintain an industrial system as required by the certificates (Case study 5).
27
Hardly any of them participated actively in the discussions, except for the executive
manager. This pattern changed only when issues were raised which were directly related to
the Chinese personnel. As a consequence of these meetings, the German management was
dissatisfied with the results, as, from a German view, there did not seem to be any decisions
made. It seemed that information transfers between Chinese managers were more likely in
small groups in a less formal surrounding. This could, in part, be explained in terms of the
concept of ‘face’ which is still important in the Chinese society. According to this, the
status of an individual within a group depends on the respect he/she receives from other
members of the group (Hwang 1987). A person can loose ‘face’ if he/she is unable or
unwilling to fulfill certain tasks and commitments to the group. Since this would cause
social isolation, Chinese workers are unlikely to admit failures openly or criticize others in
public.
It appears that German firms often send executives to their Chinese operations who are
primarily technical experts but lack knowledge on how to manage a company successfully
within an ‘inter-cultural’ context. Sometimes those managers know surprisingly little about
expectations and patterns of behavior of Chinese workers and managers. They can easily
misinterpret or misunderstand the reactions of their Chinese counterparts. Other studies
have confirmed our findings that Chinese managers in joint ventures are unsatisfied with
their position and how they are treated in many cases (Zhu, Speece and So 1998, Hoon-
Halbauer 1999). Some indicated that their experience was not respected very highly by their
joint venture partners. They also complained about the arrogance of their German
counterparts. One Chinese interviewee stated that “[i]n China market economy means
friends economy. ... you need experience not muscle!” He pointed out that German
managers who come to China are often too young and were not there long enough to know
what is going on (Interview 19). In comparing two German managers he knew really well,
another Chinese interviewee said that the one would fit better in the Chinese context and
would gain more respect by Chinese people because he made a conscious effort to listen to
the people, tried to avoid conflicts and had strict principles. The other manager, he
admitted, might have the better management skills but would not be able to mobilize
others to the same extent (Interview 15). In this case, one could say that the first manager
is more ‘powerful’ in the sense of Latour (1986), being able to mobilize people in an ‘inter-
cultural’ context (see, also, Murdoch 1995).
In joint ventures with SAIC, the German counterpart usually has a share of 50% and either
nominates the general or the deputy general manager. By contract, both sides are usually
expected to agree upon important decisions before they are implemented. But this is often
not the case in reality. Decisions are made by one side without informing the other or they
28
are dragged out (Interviews 15, 18, 33, Case studies 2, 3, 4). Problems particularly arise
when decisions have to be made about which machinery to buy, which customers to do
business with and whether certain parts of production should be outsourced. The conflicts
which arise in such situations are an important source of the development of distrust within
organizations. German executives sometimes do not know where their Chinese counterparts
are and in what kind of external meetings (i.e. within the SAIC group) they participate and
why. One interviewee mentioned that his Chinese deputy manager sometimes stays away
from the firm for a couple of days without telling anybody where he is (Case study 3).
Problems also arise between a firm’s German management and its Chinese workers. It is, for
instance, difficult to persuade the workers that production routines have to be changed if
the Chinese management does not clearly support this. It seems that allegiance towards
their Chinese bosses is greater than that towards their German ones (Case studies 2, 4). One
German interviewee thought that this was natural. If the Chinese workers would lean too
much towards the German management, they would be on their own once the Germans
eventually return to their German headquarters (Interview 33). Severe problems can arise
when German managers lose their temper in front of Chinese co-workers. According to the
concept of ‘face’ (Hwang 1987), this would be considered by the Chinese as being a loss of
self control and thus viewed negatively.
4.4 The role of boundary spanners
The above arguments have shown that ‘cultural barriers’ and embeddedness in different
institutional contexts can create massive problems in communication within joint venture
operations, as well as in relations with local suppliers. Because of these obstacles, those
people who have the capability to become network builders in both ‘cultures’ and mediate
between the German and Chinese sides are of great importance in the long-term success of
a joint venture (Bathelt, Malmberg and Maskell 2002, Coe and Bunnell 2003). These
boundary spanners might be the first ones to form a larger transnational community within
the Shanghai automobile industry cluster. The findings from our research suggest that there
are boundary spanners on each side of the joint ventures, which have been successful to
varying degrees.
(i) Chinese boundary spanners: It is often assumed that Chinese managers will increasingly
be given responsibility for the joint venture operations. To which degree this will happen is
still unclear. The managers we have met in such positions were relatively few. All of them
had spent some years in Germany and had studied or lived there (Interviews 1, 14, 43, 45).
Their advantage is that they know both languages and ‘cultural’ contexts and are able to
29
switch between different modes of interaction. Theoretically, they are extremely well suited
for establishing communication between the German and Chinese sides, solving conflicts
and circulating important knowledge. They are also able to effectively communicate with
the German headquarters.
Despite the enormous potential these boundary spanners have, their power within the joint
venture is seemingly limited. Their problem seems to be that they do not fully belong to
either side. The Chinese are skeptical that they would be too supportive to the German side,
while the Germans, especially those executives in the headquarters, do not include them in
all strategic decision-making. Some of the Chinese boundary spanners mentioned that they
wish the German headquarters would listen to them more carefully and give them more
power in decision-making (Interviews 1, 14, 15, 45).
(ii) German boundary spanners: German boundary spanners have adapted to the Chinese
business context quite well. They can often imagine to stay for a much longer time period.
Two managers married Chinese women, for instance, and have become increasingly familiar
with Chinese customs and life-styles (Interviews 2, 12). Some of these boundary spanners
have studied Sinology at a German university. Others have participated in special training
programs to prepare them for future jobs in China, consisting of a language course and work
placement for several months in a German firm in China (Interviews 5, 22, Case study 3).
Another group of Germans had no special preparation before going to China but they had
international work experience. They had proven that they are able to adjust to a new
‘cultural’ context and cope with it successfully (Interviews 16, 20). One manager of this
latter group mentioned that there was a lot of distrust between him and his Chinese
counterparts when he started his term in Shanghai. Over time, however, they began to
exchange gifts and developed close relationships. This happened because the German made
an effort to participate in Chinese social events and, among other things, invited his
partners for dinner parties at his home (Case study 3).
Not all Germans interviewed in Shanghai can be viewed as active boundary spanners. They
are under strong pressures from different sides, as they have to deal with customers which
require them to drastically cut costs. In addition, they have to act and interact within
complex Chinese settings and a very dynamic industry context. In general, we observed that
the motivation to understand and follow the rules of the Chinese side and to become an
active boundary spanner was low when the German personnel in the Shanghai operations
was relatively large or when the managers knew that they were going to leave the city soon.
30
5. Conclusions
In our analysis of the automobile industry in Shanghai, we have tried to provide evidence
that it is useful to employ a cluster concept to analyze the dynamics of production
organization in an international context. In choosing this approach, we found that it is
necessary to emphasize those cluster dimensions which are often somewhat neglected in
the cluster literature, i.e. issues related to the role of ‘culture’, institutions and power and
how they shape economic action and interaction. Through this, it was possible to apply this
approach to the ‘inter-cultural’ context of automobile production in Shanghai and
emphasize the role of policy and planning in the establishment of this industry.
Overall, the establishment of the Shanghai automobile industry cluster has been largely
planned by international automobile producers, in our case VW, making a strategic move to
extend their production system and market reach to China. As a result, a focal cluster is
developing which is characterized by a growing vertical dimension of suppliers and service
providers. These have been required by VW to establish production in China, to fulfill local-
content regulations which were, in turn, demanded by the Chinese government. The
horizontal cluster dimension is at this point still in a stage of infancy but changes quickly
as GM has also established production in the region and attracts its own suppliers.
The complex relationships between economic development and politics become obvious
through the role of SAIC, the dominant Chinese actor in the automobile industry. SAIC is
largely owned by the municipality of Shanghai and is thus shaped by the city’s general
policies and strategies with respect to future development. At the same time, SAIC is
involved in a large number of joint ventures with foreign automobile firms and parts
producers, serving to closely intertwine economic and political aspects. What makes foreign
direct investments in Shanghai’s automobile industry even more difficult is that the
conventions, rules, norms and expectations in human interaction are different from those
which the firms are used to. This, in turn, effects the organization of production and labor
processes within and between firms and the ways how firms can get knowledge about their
local markets. The successful establishment of ‘inter-cultural’ communication in the
Shanghai automobile industry cluster therefore requires the development of a shared
institutional context which enables interaction and knowledge creation. It also requires the
development of methods and routines to control production and pass on orders and
messages effectively within the production system.
We have shown how German firms import formal institutions to the Shanghai context and
begin to create new institutions to enable effective intra-firm communication and
interaction with local suppliers and markets. We argue that network builders and boundary
31
spanners, especially those originating from transnational communities, can become quite
important in overcoming problems and mediating between different positions in an ‘inter-
cultural’ context. Their ability to stimulate communication might in fact determine the
long-term success of such investments. At present time, many new operations in the
Shanghai automobile industry might still suffer from some problems and inefficiencies. This
is quickly changing, however, as Shanghai develops into an important cluster of automobile
production and firms learn to adjust to the local conditions at hand.
Acknowledgements: An earlier version of this paper was presented at the 2003 annual meeting of the Association of American Geographers in New Orleans. We would like to thank numerous participants of this meeting for their comments and suggestions. Our special thanks go to Gang Zeng for stimulating discussions on core issues of this paper and for his overwhelming hospitality during our research trips to China. We would also like to thank Weihua Je for her organizational support in Shanghai and Clare Wiseman for making her usual decisive comments at all stages of the research process. Thanks are also due to the Deutsche Forschungsgemeinschaft (German Science Foundation) for providing the funds which allowed us to conduct this research and present it to colleagues. Finally, we would like to thank those executives and managers of German and Chinese firms for their interest and time they shared with us in answering our questions. Without their support, this study would not have been possible.
32
References
Allen, J. (1997): Economies of power and space. In: Lee, R. and Wills, J. (eds.): Geographies of Economies. pp. 59-70. London: Arnold.
Bathelt, H. (2002): The re-emergence of a media industry cluster in Leipzig. European Planning Studies 10: 583-611.
Bathelt, H. and Depner, H. (2003): Innovation, Institution und Region: Zur Diskussion über nationale und regionale Innovationssysteme (Innovation, institution and region: a commentary on the discussion of national and regional innovation systems). Erdkunde 57: 126-143.
Bathelt, H. and Glückler, J. (2003): Toward a relational economic geography. Journal of Economic Geography 3: 117-144.
Bathelt, H. and Jentsch, C. (2002): Die Entstehung eines Medienclusters in Leipzig: Neue Netzwerke und alte Strukturen (The genesis of a new media industry cluster in Leipzig: new networks and old structures). In: Gräf, P. and Rauh, J. (eds.): Networks and Flows: Telekommunikation zwischen Raumstruktur, Verflechtung und Informationsgesellschaft. pp. 31-74. Hamburg, Münster: Lit.
Bathelt, H. and Taylor, M. (2002): Clusters, power and place: inequality and local growth in time-space. Geografiska Annaler 84 B: 93-109.
Bathelt, H., Malmberg, A. and Maskell, P. (2002): Clusters and Knowledge: Local Buzz, Global Pipelines and the Process of Knowledge Creation. DRUID Working Paper 2002-12. Copenhagen (URL: http://www.druid.dk/wp/wp.html).
Berndt, C. (2001): El Paso del Norte ... Modernisierungsdiskurse, Grenzziehungen und Management-Praxis in der Maquiladora-Industrie (El Paso del Norte ... Modernization utopias, ‘othering’ and management practices in Mexico’s maquiladora industry). Erdkunde 55: 244-256.
Boeckler, M. (1999): Entterritorialisierung, ‘orientalische’ Unternehmer und die diakritische Praxis der Kultur (De-territorialization, ‘oriental’ entrepreneurs and the diacritical practice of culture). Geographische Zeitschrift 87: 178-193.
Boyer, R. (1998): Hybridization and models of production: geography, history and theory. In: Boyer, R., Charron, E., Jürgens, U. and Tolliday, S. (eds.): Between Imitation and Innovation. The Transfer and Hybridization of Productive Models in the International Automobile Industry. pp. 23-56. Oxford: Oxford University Press.
Bresnahan, T., Gambardella, A. and Saxenian, A. (2001): ‘Old economy’ inputs for ‘new economy’ outcomes: cluster formation in the new Silicon Valleys. Industrial and Corporate Change 10: 835-860.
Brown, J. S. and Duguid, P. (1991): Organizational learning and communities of practice: toward a unified view of working, learning, and innovation. Organization Science 2: 40-57.
Brown, J. S. and Duguid, P. (2000): Balancing act: how to capture knowledge without killing it. Harvard Business Review 78 (May-June): 73-80.
33
Cavallardo, D. (2001): Critical and Cultural Theory: Thematic Variations. London, New Brunswick (NJ): Athlone.
China Daily (2002): SAIC buys 10% stake in GM. China Daily, 11 October: p. 5.
Coe, N. M. and Bunnell, T. G. (2003): ‘Spatializing’ knowledge communities: towards a conceptualisation of transnational innovation networks. Global Networks 3: forthcoming.
Cohen, W. M. and Levinthal, D. A. (1990): Absorptive capacity: a new perspective on learning and innovation. Administrative Science Quarterly 35: 128-152.
Cooke, P. (1999): The co-operative advantage of regions. In: Barnes, T. J. and Gertler, M. S. (eds.): The New Industrial Geography: Regions, Regulation and Institutions. pp. 54-73. London: Routledge.
Cooke, P. and Morgan, K. (1998): The Associational Economy. Oxford: Oxford University Press.
Crang, P. (1997): Cultural turns and the (re)constitution of economic geography. In: Lee, R. and Wills, J. (eds.): Geographies of Economies. pp. 3-15. London, New York: Arnold.
Depner, H. (2003): Globale Produktionsnetze und Entwicklungsländer. Die Automobilindustrie in China (Global production networks and development countries. The case of the automobile industry in China). Praxis Geographie 33 (2), 22-25.
Dicken, P., Kelly, P. F., Olds, K. and Yeung, H. W.-c. (2001): Chains and networks, territories and scales: towards a relational framework for analysing the global economy. Global Networks 1: 89-112.
Fujita, M., Krugman, P. and Venables, A. J. (1999): The Spatial Economy. Cities, Regions and International Trade. Cambridge (MA): MIT Press.
Gaebe, W. (1993): Neue räumliche Organisationsstrukturen in der Automobilindustrie (New spatial organization of the automobile industry). Geographische Rundschau 45: 493-497.
Gereffi, G. (1994): The organization of buyer-driven global commodity chains: how US retailers shape overseas production networks. In: Gereffi, G. and Korzeniewicz, M. (eds.): Commodity Chains and Global Capitalism. pp. 95-122. Westport (CT): Praeger.
Gereffi, G. (1999): International trade and industrial upgrading in the apparel commodity chain. Journal of International Economics 48: 37-70.
Gereffi, G. and Korzeniewicz M. (1990): Commodity chains and footwear exports in the semiperiphery. In: Martin, W. G. (ed.): Semiperipheral States in the World-Economy. pp. 45-68. Westport (CT): Greenwood Press.
Gertler, M. S. (1993): Implementing advanced manufacturing technologies in mature industrial regions: towards a social model of technology production. Regional Studies 27: 665-680.
Gertler, M. S. (1995): ‘Being there’: proximity, organization, and culture in the development and adoption of advanced manufacturing technologies. Economic Geography 71: 1-26.
34
Gertler, M. S. (1997): The invention of regional culture. In: Lee, R. and Wills, J. (eds.): Geographies of Economies. pp. 47-58. London, New York: Arnold.
Gertler, M. S. (2001): Best practice? Geography, learning and the institutional limits to strong convergence. Journal of Economic Geography 1: 5-26.
Giuliani, E. (2002): Cluster Absorptive Capability: An Evolutionary Approach for Industrial Clusters in Developing Countries. Paper presented at the DRUID Summer Conference on ‘Industrial Dynamics of the New and Old Economy -- Who is Embracing Whom?’ in Copenhagen & Elsinore (URL: http://www.druid.dk/conferences/summer2002/Papers/GIULIANI.pdf).
Gordon, I. R. and McCann, P. (2000): Industrial clusters: complexes, agglomeration and/ or social networks. Urban Studies 37: 513-532.
Grabher, G. (2001): Ecologies of creativity: the village, the group, and the heterarchic organisation of the British advertising industry. Environment and Planning A 33: 351-374.
Grabher, G. (2002a): Cool projects, boring institutions: temporary collaboration in social context. Regional Studies 36: 205-214.
Grabher, G. (2002b): The project ecology of advertising: tasks, talents and teams. Regional Studies 36: 245-262.
Granovetter, M. (1973): The strength of weak ties. American Journal of Sociology 78: 1360-1380.
Harwit, E. (1992): Foreign passenger car ventures and Chinese decision making. The Australian Journal of Chinese Affairs 28, 141-166.
Henderson, J., Dicken, P., Hess, M., Coe, N. and Yeung, H. W.-c. (2002): Global production networks and the analysis of economic development. Review of International Political Economy 9: 436.464.
Hodgson, G. M. (1988): Economics and Institutions: A Manifesto for a Modern Institutional Economics. Cambridge: Polity Press.
Hoon-Halbauer, S. K. (1999): Managing relationships within Sino-foreign joint ventures. Journal of World Business 34, 344-371.
Hösle, V. (2002): Was ist Kultur? (What is culture?) Think on 1: 12-19.
Hsu J.-y. and Saxenian, A. (2000): The limits to guanxi capitalism: transnational collaboration between Taiwan and the US. Environment and Planning A 32: 1991-2005.
Humphrey, J. and Schmitz, H. (2002): How does insertion in global value chains affect upgrading in industrial clusters? Regional Studies 36: 1017-1027.
Hwang, K.-k. (1987): Face and favor: the Chinese power game. American Journal of Sociology 92, 944-974.
Kern, H. (1996): Vertrauensverlust und blindes Vertrauen: Integrationsprobleme im ökonomischen Handeln (Loss of trust and blind confidence in economic action). SOFI-Mitteilungen 24/1996: 7-14.
35
Kiefer, T. (1998): Volkswagen’s Shanghai plant: between Chinese tradition and modernization strategy. In: Boyer, R., Charron, E., Jürgens, U. and Tolliday, S. (eds.): Between Imitation and Innovation. The Transfer and Hybridization of Productive Models in the International Automobile Industry. pp. 342-360. Oxford: Oxford University Press.
Krugman, P. (1991): Geography and Trade. Leuven: Leuven University Press & Cambridge (MA), London: MIT Press.
Krugman, P. (2000): Where in the world is the ‘new economic geography’? In: Clark, G. L., Feldman, M. P. and Gertler, M. S. (eds.): The Oxford Handbook of Economic Geography. pp. 49-60. Oxford: Oxford University Press.
Lash, S. and Urry, J. (1994): Economies of Signs and Spaces. London: Sage.
Latour, B. (1986): The powers of association. In: Law, J. (ed.): Power, Action and Belief: A New Sociology of Knowledge? pp. 264-280. London: Routledge & Kegan Paul.
Lawson, C. (1999): Towards a competence theory of the region. Cambridge Journal of Economics 23: 151-166.
Lawson, C. and Lorenz, E. (1999): Collective learning, tacit knowledge and regional innovative capacity. Regional Studies 33: 305-317.
Lee, C., Chen, J. and Fujimoto, T. (1996): Different Strategies of Localization in the Chinese Auto Industry: The Cases of Shanghai Volkswagen and Tianjin Daihatsu. Working Paper for the MIT 1996 IMVP Sponsors Meeting in Sao Paulo.
Lorenz, E. (1999): Trust, contract and economic cooperation. Cambridge Journal of Economics 23: 301-315.
Lundvall, B.-Å. (1988): Innovation as an interactive process: from user-producer interaction to the national system of innovation. In: Dosi, G., Freeman, C., Nelson, R. R., Silverberg, G. and Soete, L. L. G. (eds.): Technical Change and Economic Theory. pp. 349-369. London: Pinter.
Lundvall, B.-Å. and Maskell, P. (2000): Nation states and economic development - From national systems of production to national systems of knowledge creation and learning. In: Clark, G. L., Feldman, M. P. and Gertler, M. S. (eds.): The Oxford Handbook of Economic Geography. pp. 353-372. Oxford: Oxford University Press.
Maillat, D. (1998): Vom ‘Industrial District’ zum innovativen Milieu: Ein Beitrag zur Analyse der lokalen Produktionssysteme (From industrial networks to innovative milieus: towards an analysis of territorial production systems). Geographische Zeitschrift 86: 1-15.
Malecki, E. J. (2000): Knowledge and regional competitiveness. Erdkunde 54: 334-351.
Malmberg, A. and Maskell, P. (2002): The elusive concept of localization economies: towards a knowledge-based theory of spatial clustering. Environment and Planning A 34: 429-449.
Marshall, A. (1920): Principles of Economics. 8th edition. Philadelphia: Porcupine Press.
Marshall, A. (1927): Industry and Trade. A Study of Industrial Technique and Business Organization; and Their Influences on the Conditions of Various Classes and Nations. 3rd edition. London: Macmillan.
36
Martin, R. and Sunley, P. (2003): Deconstructing clusters: chaotic concept or policy panacea? Journal of Economic Geography 3: 5-35.
Maskell, P. (2001): Towards a knowledge-based theory of the geographical cluster. Industrial and Corporate Change 10: 921-943.
Maskell, P. and Malmberg, A. (1999a): The competitiveness of firms and regions: ‘ubiquitification’ and the importance of localized learning. European Urban and Regional Studies 6: 9-25.
Maskell, P. and Malmberg, A. (1999b): Localised learning and industrial competitiveness. Cambridge Journal of Economics 23: 167-185.
Maskell, P., Eskelinen, H., Hannibalsson, I., Malmberg, A. and Vatne, E. (1998): Competitiveness, Localised Learning and Regional Development: Specialisation and Prosperity in Small Open Economies. London, New York: Routledge.
McKelvey, M. (1997): Using evolutionary theory to define systems of innovation. In: Edquist, C. (ed.): Systems of Innovation: Technologies, Institutions and Organizations. pp. 200-222. London: Pinter.
Murdoch, J. (1995): Actor-networks and the evolution of economic forms: combining description and explanation in theories of regulation, flexible specialization, and networks. Environment and Planning A 27: 731-757.
North, D. C. (1990): Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press.
Oinas, P. (1997): On the socio-spatial embeddedness of business firms. Erdkunde 51: 23-32.
Owen-Smith, J. and Powell, W. W. (2002): Knowledge Networks in the Boston Biotechnology Community. Paper presented at the Conference on ‘Science as an Institution and the Institutions of Science’ in Siena.
Porter, M. E. (1990): The Competitive Advantage of Nations. New York: Free Press.
Porter, M. E. (1998): Clusters and the new economics of competition. Harvard Business Review 76 (November-December): 77-90.
Porter, M. E. (2000): Locations, clusters, and company strategy. In: Clark, G. L., Feldman, M. P. and Gertler, M. S. (eds.): The Oxford Handbook of Economic Geography. pp. 253-274. Oxford: Oxford University Press.
Power, D. (2002): ‘Cultural industries’ in Sweden: an assessment of their place in the Swedish economy. Economic Geography 78: 103-127.
Roland Berger Strategy Consultants (2001): 10 Mega Trends of the Chinese Automotive Industry after WTO Entry. Roland Berger Study. Shanghai: Roland Berger Strategy Consultants.
SAIC - Shanghai Automotive Industry Corporation (2002): Annual Report 2001. Shanghai.
Saxenian, A. and Hsu J.-y. (2001): The Silicon Valley-Hsinchu connection: technical communities and industrial upgrading. Industrial and Corporate Change 10: 893-920.
Sayer, A. (1997): The dialectic of culture and economy. In: Lee, R. and Wills, J. (eds.): Geographies of Economies. pp. 16-26. London, New York: Arnold.
37
Schamp, E. W. (1995): The German automobile production system going global. In: Hudson, R. and Schamp, E. W. (eds.): Towards a New Map of Automobile Manufacturing in Europe? New Production Concepts and Spatial Restructuring. pp. 93-116. Berlin, Heidelberg: Springer.
Schramm, M. and Taube, M. (2002): Erfolg ohne Guanxi? (Success without guanxi?) China Contact 2/2002: 44-46
Scott, A. J. (1988): New Industrial Spaces: Flexible Production Organization and Regional Development in North America and Western Europe. London: Pion.
Scott, A. J. (1998): Regions and the World Economy: The Coming Shape of Global Production, Competition, and Political Order. Oxford, New York: Oxford University Press.
Shanghai Daily (2002): Chinese-car sales rise 40%. Shanghai Daily, 10 September: p. 2.
Sit, V. F. S. and Liu, W. (2000): Restructuring and spatial change of China’s auto industry under institutional reform and globalization. Annals of the Association of American Geographers 90, 653-673.
Smith, A. (2003): Power relations, industrial clusters, and regional transformations: pan-European integration and outward processing in the Slovak clothing industry. Economic Geography 79: 17-40.
Sofer, M. and Schnell, I. (2002): Over- and under-embeddedness: failures in developing mixed embeddedness among Israeli Arab entrepreneurs. In: Taylor, M. and Leonard, S. (eds.): Embedded Enterprise and Social Capital: International Perspectives. pp. 207-224. Aldershot: Ashgate.
Storper, M. (1995): The resurgence of regional economics, ten years later. European Urban and Regional Studies 2: 191-221.
Storper, M. (1997): The Regional World. Territorial Development in a Global Economy. New York, London: Guilford.
Storper, M. and Venables, A. J. (2002): Buzz: The Economic Force of the City. Paper presented at the DRUID Summer Conference on ‘Industrial Dynamics of the New and Old Economy -- Who is Embracing Whom?’ in Copenhagen & Elsinore.
Storper, M. and Walker, R. (1989): The Capitalist Imperative. Territory, Technology, and Industrial Growth. New York, Oxford: Basil Blackwell.
Taylor, M. (2000): Enterprise, power and embeddedness: an empirical exploration. In: Vatne, E. and Taylor, M. (eds.): The Networked Firm in a Global World: Small Firms in New Environments. pp. 199-233. Aldershot, Burlington: Ashgate.
Thrift, N. (2000): Pandora’s box? Cultural geographies of economies. In: Clark, G. L., Feldman, M. P. and Gertler, M. S. (eds.): The Oxford Handbook of Economic Geography. pp. 689-704. Oxford: Oxford University Press.
Tracey, P., Clark, G. L., and Lawton Smith, H. (2002): Cognition, Learning and European Regional Growth: An Agent-Centred Perspective on the ‘New’ Economy. Paper presented at the Annual Meeting of the Association of American Geographers in Los Angeles.
Uzzi, B. (1997): Social structure and competition in interfirm networks: the paradox of embeddedness. Administrative Science Quarterly 42: 35-67.
38
van den Bosch, F. A. J., Volberda, H. W. and de Boer, M. (1999): Coevolution of firm absorptive capacity and knowledge environment: organizational forms and combinative capabilities. Organizational Science 10: 551-568.
Veloso, F. and Kumar, R. (2002): The Automotive Supply Chain: Global Trends and Asian Perspectives. ERD Working Paper Series 3. Asian Development Bank.
Vwd: Asien (2003a): GM weitet Produktion aus (GM extends production). vwd: asien, 1-2/2003: p. 2.
Vwd: Asien (2003b): GM: Kfz-Absatz verdreifacht (GM: car sales have tripled). vwd: asien, 4/2003, p. 4.
Wang, H. (2001): Weak State, Strong Networks. The Institutional Dynamics of Foreign Direct Investment in China. Oxford: Oxford University Press.
Wenger, E. (1998): Communities of Practice: Learning, Meaning, and Identity. Cambridge: Cambridge University Press.
Wenger, E. C. and Snyder, W. M. (2000): Communities of practice: the organizational frontier. Harvard Business Review 78 (January-February): 139-145.
Yang, X. (1995): Globalization of the Automobile Industry. The United States, Japan and the People’s Republic of China. Westport (CT): Praeger.
Yeung, H. W.-c. (2000): Embedding foreign affiliates in transnational business networks: the case of Hong Kong firms in Southeast Asia. Environment and Planning A 32: 201-222.
Zeng, G. (2000): The financial crisis in Asia and the modification of economic structure in Shanghai. Asian Geographer 19: 37-48.
Zhu, G., Speece, M. W. and So, S. L. M. (1998): Conflicts in Sino-European joint ventures. In: Selmer, J. (ed.): International Management in China. pp. 13-28. London: Routledge.